CoMAT: Chain of Mathematically Annotated Thought Improves Mathematical Reasoning
This addresses the problem of improving mathematical reasoning for users of LLMs, representing an incremental advancement over existing prompting techniques.
The paper tackles the challenge of mathematical reasoning in large language models by introducing CoMAT, a method that converts natural language queries to symbolic form and executes reasoning, resulting in performance gains such as 4.48% on MMLU-Redux and 4.58% on GaoKao MCQ benchmarks.
Mathematical reasoning remains a significant challenge for large language models (LLMs), despite progress in prompting techniques such as Chain-of-Thought (CoT). We present Chain of Mathematically Annotated Thought (CoMAT), which enhances reasoning through two stages: Symbolic Conversion (converting natural language queries into symbolic form) and Reasoning Execution (deriving answers from symbolic representations). CoMAT operates entirely with a single LLM and without external solvers. Across four LLMs, CoMAT outperforms traditional CoT on six out of seven benchmarks, achieving gains of 4.48% on MMLU-Redux (MATH) and 4.58% on GaoKao MCQ. In addition to improved performance, CoMAT ensures faithfulness and verifiability, offering a transparent reasoning process for complex mathematical tasks